browser-use

Web scraping with browser automation navigates pages, extracts text and links, and saves outputs using Playwright. This Skill handles navigation, selector-based extraction, and content capture across diverse websites without manual browsing.

Authentication-aware workflows create and reuse persistent login sessions via Playwright, enabling automated access to protected sites. Store account credentials once and reuse them across multiple authenticated requests without repeated logins.

Browser automation captures screenshots, saves PDFs, and downloads files programmatically. Use built-in commands to interact with web content, trigger downloads, and automate file uploads across any website.

Browser automation with Playwright executes JavaScript, handles dynamic content, and manages interactive elements like forms and clicks. Headless or headful modes support both manual and script-driven workflows that simple HTTP requests cannot replicate.

Headless and headful operation modes are both supported for script-driven automation. Headless runs efficiently in production environments; headful enables interactive debugging and manual session inspection when needed.

Playwright is a core dependency that handles browser control, navigation, and interaction. This Skill manages Playwright execution through the uv CLI, so you invoke commands without directly managing the browser instance.